This monograph, which uses Conversation Analysis (CA) as a framework to show how issues of intersubjectivity (i.e. mutual understanding) are managed in French talk-in-interaction, is the author's doctoral dissertation undertaken at Lund University. It is printed unrevised, in line with Swedish academic tradition. Fortunately, the lack of revision has no noticeable bearing on what is a well written and engaging thesis by a self-confessed CA ‘fundamentalist’ (v).

In his methodology chapter (Chapter Two, 5–48), itself preceded by a brief introduction (Chapter One, 1–4), Persson presents a comprehensive review of the current CA literature, together with a detailed overview of the principles of CA and an explanation of the corpora of authentic data used for the study (mostly telephone conversations). In addition to closely following traditional micro-level CA analysis, the author emphasises his innovative approach of incorporating an auditive and acoustic prosodic analysis; he is one of the first to do this with French data. Persson examines three conversational practices combining linguistic resources, sequential positioning and prosody which may be used when intersubjectivity in interaction, for the most part implicit or implied, is not immediately attained, viz. formulations (Chapter Three, 49–105), ah-prefaced other-repeats (Chapter Four, 107–181) and simple other-repeats (Chapter Five, 183–241).

The chapter on formulations (where the speaker summarises and/or reformulates what has been said to demonstrate understanding of the interlocutor's earlier utterance) reviews previous research and shows how formulations differ from repairs and ‘recipient-oriented topic proffers’ (Schegloff). Persson presents a fine-grained interactional and prosodic analysis of several extracts of data, illustrating how interlocutors orient towards the final accented syllable of the formulation as a transition relevance place (where the turn may go to the interlocutor or the current speaker may continue) and showing how contrasting prosody is treated as significant by the interlocutors and how this affects the subsequent interaction. Whereas formulations with a final rising intonation invite elaborate confirmation (e.g. a confirmation followed by a justification, explanation or more details) and therefore extend the interactional sequence, formulations with a final rise-fall intonation induce a simple confirmation from the interlocutor, often corresponding to making arrangements and closing the conversation or a sequence of interactional trouble.

Next in line are ‘ah-prefaced other-repeats’, where the reception marker ah is followed by a partial or complete repetition of the preceding turn, without pause. Persson shows how, contrary to claims made in some of the non-CA literature, these markers differ from simple backchannels or other signs of confirmation that the listener has heard or understood the previous utterance. The ah-prefaced other-repeat acknowledges the interlocutor's utterance while also marking the speaker's own preceding utterance as inadequate (e.g. inappropriate or incorrect) in some way. It then reorients the subsequent conversation based on the new knowledge. A large part of the chapter deals with the positioning of the ah-prefaced other-repeat in the adjacency pair, with a range of ‘attendant activities’ (134) which can accompany the reception marker such as an apology, a justification etc., and with extracts where the ah-prefaced other-repeat would not be possible. Persson then presents the phonetic properties and functions of the ah-prefaced other-repeat, of which there are two main contrasting variants: strong and weak. The strong variant (with an emphasis on the final syllable) treats the utterance as not completely accepted or as problematic, while the weak variant (with a falling or flat intonation on the final syllable) accepts the received utterance as complete or unproblematic. Persson acknowledges that there are some problems with what he calls an ‘approximate, even simplified’ phonetic analysis (180), suggesting that further work remains to be done.

Chapter Five examines how simple other-repeats are understood by the interlocutor, both interactionally and phonetically. Following a review of the literature on other-repeats, Persson presents a number of examples which illustrate how the positioning of the other-repeat affects the subsequent interaction, including overlapping talk. The subsequent detailed phonetic analysis of several extracts (which in fact constitutes almost the whole chapter) shows how secondary emphasis and falling intonation point to a lack of necessity of reply (to the repeat), but how primary emphasis with rise-fall intonation requires further elaboration, indicating interactional trouble. The final section of the chapter deals with weak and strong confirmative responses and illustrates how the interactional work achieved by these responses is inherently tied to their linguistic and phonetic form. Persson demonstrates how weak confirmative responses are generally produced in answer to repeats with minimal expansion, while strong confirmative responses are produced in reaction to repeats which form the first part of a non-minimal expansion, i.e. they signal interactional trouble of some kind. The chapter ends with an interesting discussion on the weak versus strong phonetic variants of the confirmative particle oui/ouais, followed by a final section on atypical confirmative responses.

Chapter Six (243–263) summarises the study and reiterates the important contribution of the phonetic analysis, while acknowledging certain limitations of the research. It is here that more discussion around the corpora used for this study would have been appropriate. Scant information is supplied in the methodology chapter and in an appendix, neither of which include the dates of the recordings, the age, gender or socio-economic background of the speakers, nor the types of French analysed (although it is noted that one corpus includes ‘plusieurs variations européennes de français’, 289). As a CA ‘fundamentalist’, Persson probably had no qualms about disregarding external evidence, since according to strict CA principles everything relevant for the analysis is contained in the data. Nevertheless, some discussion around the inclusion of different varieties of French would have been particularly relevant, especially in view of the importance placed on the phonetic analysis of the data. Did all speakers display similar prosody and intonation? Can the author confirm that the different varieties of French had no bearing on the findings?

Apart from the above, which is arguably more of an inherent flaw in CA methodology than a weakness in Persson's thesis, this study makes a significant, original and valuable contribution to the existing literature on French talk-in-interaction and to intersubjectivity in particular. Not only does the author meticulously analyse numerous extracts of interaction to demonstrate how the three conversational practices under examination are not just instances of repair or backchannels, he presents a convincing case for the inclusion of prosodic information as a valuable tool in our understanding of social interaction. Persson repeatedly points out that the combination of prosody and sequential positioning determines the meaning of any conversational practice, and that neither can be considered in isolation to mean the same thing. Future research in talk-in-interaction will benefit from this combined approach.